Re: [syzbot] [bcachefs?] [rcu?] BUG: unable to handle kernel NULL pointer dereference in rcu_core (3)

[Boqun Feng]

On Thu, Jun 12, 2025, at 2:37 AM, Uladzislau Rezki wrote:
> On Thu, Jun 12, 2025 at 09:42:32AM +0200, Aleksandr Nogikh wrote:
>> On Wed, Jun 11, 2025 at 10:58 PM Boqun Feng <boqun.feng@xxxxxxxxx> wrote:
>> >
>> > On Wed, Jun 11, 2025 at 12:57:04PM -0700, syzbot wrote:
>> > > Hello,
>> > >
>> > > syzbot has tested the proposed patch and the reproducer did not trigger any issue:
>> > >
>> > > Reported-by: syzbot+80e5d6f453f14a53383a@xxxxxxxxxxxxxxxxxxxxxxxxx
>> > > Tested-by: syzbot+80e5d6f453f14a53383a@xxxxxxxxxxxxxxxxxxxxxxxxx
>> > >
>> > > Tested on:
>> > >
>> > > commit:         488ef356 KEYS: Invert FINAL_PUT bit
>> > > git tree:       upstream
>> > > console output: https://syzkaller.appspot.com/x/log.txt?x=129a660c580000
>> >
>> > Is there a way to see the whole console output? If Ulad's patch fixes
>> > the exact issue, we should be able to see a WARN_ON_ONCE() triggered.
>> 
>> If WARN_ON_ONCE() were triggered, the associated kernel panic output
>> would have been at the end of this log.
>> 
>> >
>> > Regards,
>> > Boqun
>> >
>> > > kernel config:  https://syzkaller.appspot.com/x/.config?x=713d218acd33d94
>> > > dashboard link: https://syzkaller.appspot.com/bug?extid=80e5d6f453f14a53383a
>> 
>> FWIW the last time the bug was observed on syzbot was 100 days ago, so
>> it has likely been fixed since then or has become much harder to
>> reproduce.
>> 
> That is even worse, if it is last for 100 days already.
>

My understanding is that the evidence shows that the
issue that directly caused null-ptr-derek the has been
fixed 100 days ago.

Regards,
Boqun

> --
> Uladzislau Rezki